RISK BASED ASSET INTEGRITY INDICATORS
by
© MD. JAKIUL HASSAN
A Thcsis submittcd to thc School ofGraduatc Studies in partial fulfillmcntofthe rcquircmcnts forthcdcgrccof
MASTER OF ENGINEERING
Faculty ofEnginccring and Applied Scicncc Memorial Universityof Newfoundland
September 2011
St. John's Newfoundland
ABSTRACT
Currently, asset integrity is a major concern and presents challenges to the process industry that cannot bc ignored. Asscssing asset performancc is also a difficult task, due to the involvement of versatile tangible, as well as intangible, assets' performance measuring parameters. Monitoring and assessing asset performance through indicators is gaining popularity in several sectors. However, the lack of a comprehensive set of appropriate indicators' development strategy, quantification technique, and measurement cohesion limit the usc of an indicator system. To overcome these problems, a hierarchical fi'amework is developed for identifying indicators and monitoring the performance of the asset. The hierarchical structure attempts to characterize the asset and relate it10a company's strategic goal. The hierarchical structure is based on the three major areas of asset integrity, and provides an opportunity to follow bottom-up perspective for identifying multilevel level indicators. This approach uses a risk metric to classify asset integrity, and risk provides a common ground to integrate leading and lagging indicators.
The hierarchical structure is followed because the specific indicator results will have no values unless they are linked to the ultimate goal for ensuring asset integrity by measuring asset performance. Similarly, this framework and indicator will have no values unless a mathematical model is used to quantifY the risk information. The analytical hierarchy process is used to determine the weight or prioritization of each level indicator and the aggregation of the indicators' outcomes arc done depending on the associated risk. This will eventually aid in assessing asset risk based performance. To validate the developed model and to quantify the condition of assets ofa process plant a benchmark study is
conductcd. Thc estimated index value will determine the condition of the asset based on the performance risk index scale. As a result, the indicator system can provide a comprehensive view on a process plant equipment status and also can Icad to the particular consideration of trends requiringattcntion.
ACKNOWLEDGEMENTS
The first and foremost person to whom I would like to express my sincere gratitude is my supervisor,Dr. Faisal Khan, for his incessant encouragement, motivation, guidance, efforts, and excellent mentoring.Ido acknowledge his support and patience throughout the entire path of pursuing this degree.
Iwould also like to express my earnest gratitude to the personnel who have participated and given valuable feedback for the benchmark study and pair-wise importance comparison questionnaire. They have sacrificed their valuable time, shared information, and deliberated efforts, for whichIam very grateful.
Iwould like to extend my acknowledgement to my colleagues in the process simulation laboratory for support and valuable advice.Inparticular, Iwould like to (hank Premkumar Thodi, Refaul Ferdous and Sikdar Mainul Hassan for their encouragement through discussion, invaluable suggestions, and beneficial assistance.
Furthermore,Iwould like to thank my family: my parents, especially my mother, parents in law, brother, and sister for their inspiration and endless support.
Finally,Iam very grateful to my wife Rumna Afroz and acknowledge her sacrifices.
Without her support it would not be possible for me to complete the research work.Iam absolutelyhumbledbyherexceptionaldevotionandeternalmentalsupport.
iv
Table of Contents
ABSTRACT.
ACKNOWLEDGEMENTS List of Tables.
List of Figures
List of Abbreviations and Symbols List of Appendices
.... ii
... iv
... xiii
Chapter I: Introduction I 1.4 Factors Affecting the Integrity of Asset 1.5 Why eedPerformanceMeasurement?. 1.6 Performance Measure 1.7 Asset Integrity Indicator. ... 1.7.1 Purpose of asset integrity indicators 1.7.2 Types of Asset Integrity Indicators .. 1.7.3 Characteristics of Asset Integrity Indicators 1.8 Objectives of Research ... 1.9 Novelty of this Research ... 1.1 1.2 1.3 Background Asset Integrity . ...•.•.•. 1
... 9
... 11
... 13
.... 14
... 17
... 18
... 19
... 21
... 23
... 24
1.10 Thesis Outline ... 24
Chapter 2: Literature Overview 26
2.1 Introduction 26
2.2 Performance Measurement Approach... . 26
2.2.1 Kaplan and Norton's Balanced Scorecard Approach. ... 27
2.2.2 Wireman's Hierarchical Approach . . 28
2.3 Regulatory Organization's Guidelines for Indicator Development 29 2.4 Asset Integrity Development Guideline .... ... 33 2.4.1 OGP Guideline on Asset Integrity. ... 33
2.4.2 HSE on Asset Integrity ... 34
2.5 Integrity Indicator Development Approach .... 35
2.6 Asset condition index. ... 36
2.7 Discussion and Remark .... 37
Chapter 3: Research Methodology 40
3.1 Introduction
3.2 Risk Based Asset Integrity Indicators Methodology 3.2.1 Delimitation of Asset Integrity...
3.2.1.1 Asset IntegrityElcment Interrelation ..
3.2.1.2 Major Element Contribution to Asset Integrity
... .40 ..40
. .41
. 44
... .45 3.2.2 Hierarchical Framework Development for Asset Integrity. ... .46
vi
3.2.2.1 Description of Different Stages of Indicators 3.2.3 Reason for Hierarchical Framework .
... .49
. 51
3.2.4 IndicatorFramcworkOvcrvicw 52
3.2.5 Risk Based Approach for Asset Integrity Indicator... .. 52
3.2.5.1 Risk Based Leading Indicator 54 3.2.5.2 Risk Based Lagging Indicator. 3.2.6 Performance Index Development. 3.2.7 Risk Based Indicator Development 3.2.7.1 Mechanical Integrity Indicator Development 3.2.7.2 Operational Integrity Indicator Development 3.2.7.3 Personnel Integrity Indicator Development 3.2.8 Aggregation Technique .. 3.2.9 Indicator Risk Determination 3.2.9.1 Leading Indicator Risk Determination 3.2.9.2 Lagging Indicator Risk Detemlination 3.2.10 Analytical Hierarchy Process Technique. 3.2.10.1 Multilevel Weight Calculation using AH P 3.2.10.2 Consistency Index and Consistency Ratio 3.2.11 Leading and Lagging Risk Index Scale 3.2.12 DataCollection 3.2.13 Multilevel Indicator Weight Assessment 3.2.14 Implementation of Indicators ... 55
... 56
... 56
... 58
... 68
... 78
.. 88
... 89
... 90
... 92
... 94
... 95
... 98
... 101
... 102
... 103
.106 Chapter 4: Benchmark Study 109 vii 4.1 Introduction ... 109
4.2 Qucstionnairc Dcvclopmcnt for Spccific Indicators 110
4.3 Rcsult Analysis 111
4.4 Sensitivity Analysis.... . 114
4.4.1 Impact of Indicators Wcight on Overall Risk Indcx Valuc . 115 4.4.2 Model Responsc to Changing Indicators Inputs .. 120 Chapter 5: Additional Work: Risk Based Spare Parts Inventory Management... 121
5.1 5.2
Prcamblc.
Introduction
. 121 ... 122 5.3 Criticality ranking ofcomponcnts
5.4 Sparc Parts Dcmand ForccastingTechniquc 5.5 Thc Bayesian Analysis Approach 5.6 Risk Estimation for Spare Parts Scrvicc Rcliability.
5.7 Risk Levcl Rcduction and Procurcmcnt Policy 5.8 Inspcctionlntcrval
5.9 Cost Consideration for Inspcction 5.10 Illustrativc Example 5.11 Conclusion
... 127 ... 128 ... 130 . ... 135 ... 136 .... 137 ... 138 ... 139 ... 149 Chapter 6: Summary, Conclusion and Future Research Suggestions 151
6.1 Summary
6.2 Conclusion ...
6.3 Future Rcscarch Suggestions ..
viii
... 151
. 154
. 155
Bibliography
Bibliography for Additional Work Appendixes.
... 156 ... 164 ... 168
List of Tables
Table 3-1: Random consistency indcx 101
Table 3-2: Asset risk indcx characterization scale and color code 102 Table 3-3: Standardized weights for multilevel indicator in asset indicator hierarchy .. 104 Table 4-1: Asset integrity leading and lagging risk index for participants' process plant. III Table 4-2: Sensitivity analysis scenarios with Icadingand lagging risk index outcome. 116 TableS-I: Parameter of prior distribution and spares quantity correspondstoservice level 142 Table 5-2: Risk determination based on the spare parts adequacy level lor one year 142 Table 5-3: Initial spare parts requirement quantity determination... .. 143 Table 5-4: Risk Determination following minimum replenishment time interval 143 Table 5-5: Sparc parts quantity&risk level corresponds to spare supply lead time 143
List of Figures
Figure I-I: (a) Total recordable incident rate (b) Lost time injury frequency
Figurel-2: Three types of performance measure IU
Figure 2-1: The hierarchical top down 'pertormance indicators system' 29 Figure 3-1: Methodology for estimat ing risk based asset integrity level .42 Figure 3-2: Elements that have an impact on the integrity of an asset over its life cycle .44 Figure 3-3: Relation between asset integrity with its contributOlY clement Figure 3-4: Major clement contribution to Asset Integrity performance
.... 45 ... 46 Figure 3-5: Hierarchical indicator pyramid for monitoring integrity level of asset .48 Figure 3-6: Hierarchical rramework for development of asset integrity indicators 59 Figure 3-7: Tree diagram for mechanical integrity indicators development Figure 3-8: Leading and lagging indicators tor inspection activity ..
Figure 3-9: Leading and lagging indicators for maintenance activity
... 63
. 64
... 65 Figure 3-1 0: Leading and lagging indicators for inspection and maintenance ITk'lI1agement activity 66 Figure 3-11: Leading and lagging indicators for engineering assessment activity 67 Figure 3-12: Tree diagram for operational integrity indicators development... . 72 Figure 3-13: Leading and lagging indicators for operating perfiJrmanee activity 73 Figure 3-14: Leading and lagging indicators for state ofstl1lctures, systems and components activity 74 Figure 3-15: Leading and lagging indicators for plant configuration and modification activity 75 Figure 3-16: Leading and lagging indicators for engineering safety system activity 76 Figure 3-17: Leading and lagging indicators for emergency management activity 77 Figure 3-18: Tree diagram for personnel integrity indicators development 83 xi
Figure 3-19: Leading and lagging indicators for training activity . Figure 3-20: Leading and lagging indicators for staff competence activity Figure 3-21: Leading and lagging indicators for permit to work activity Figure 3-22: Leading and lagging indicators for communication activity
... 84
... 85
... 86
... 87
Figure 3-23: Implementation of indicator system for achieving asset integrity 108 Figure 4-1: Overall asset risk index for process plants in benchmark study... .. 112
Figure 4-2: Leading risk index variation [Tom baseline risk index value 117 Figure 4-3: Leading indicator sensitivity to parameter weight in the RI outcome... . 117
Figure 4-4: Lagging risk index variation from baseline risk index value ... . 118
Figure 4-5: Lagging indicator sensitivity to parameter weight in the RJ outcome 119 Figure 5-1: Methodology for Risk-based spare parts inventory management... . .. 126
Figure 5-2: Risk level corresponds to spare parts service inadequacy level... .. ... 144
Figure 5-3: Spare parts quantity with minimum lead time interval procurement strategy. 146 Figure 5-4: Risk level variation thIOughouttheobservation period .... . 146 Figure 5-5: Optimum inspection interval that maximizes the availability of spares parts. 147
xii
List of Abbreviations and Symbols
AI Asset Integrity
MI Mechanical Integrity 01 Operationallntcgrity PI Pcrsonncllntcgrity
EI Elcmcnt Indicator
Acl Activity Indicator
KI Kcy Indicator
SI Spccific Indicator All Assetlntcgritylndicator KPI KcyPcrformancc Indicator HSE Hcalth and Safcty Exccutivc HSL Health and Safcty laboratory DNV Oct Norske Vcritas OGP Oil and Gas Producer CCPS Ccntcr for Chcmical Process Safcty
OECD Thc Organization for Economic Co-operation and Dcvclopmcnt LPG Liqucfied Petrolculll Gas
IAEA International Atomic Energy Agcncy AHP Analytical Hierarchy Proccss
MOC Management ofChangc
SSC Systcms, Structure and Componcnts xiii
WO Work Order
PTW Pcrmit to Work
LOTO Lock outs Tag outs
WA 0 World Association of uclcarOpcrators SE UF Safcty of Eastcrn Europcan Typc uclcar Facilitics SPI Sa(clyPcrl<mnancclndicator
DOE Dcpartmcnt ofEncrgy EEA Europcan Environmcnt Agcncy ARIS Assct Risk Indcx Scalc
BLEVE Boiling Liquid Expanding Vapor Explosion
W, Wcightoflndicator
Eigcnvaluc Maximum Eigcnvaluc RLDi
RLGi
xiv
Risk factor for leading indicator Risk factor for lagging indicator Indication of inconsistency of cxpcrt judgcmcnt
List of Appendices
Appendix A Consequence Class Rating... . 168
Appendix B: Questionnaires for data collection and pair-wise comparison... . 169 AppendixC:List of Experts provided feedback on pair-wise comparison 200 Appendix D: List of process plant paI1icipate in benchmark study 201
Appendix E: Definitions... . 202
Chapter 1 Introduction
1.1 Background
In the process industry, asset integrity is presently a major concern. A large number of accidents/incidents took place in process industries in the past where the failures of equipment were found to be fundamental contributory issues. Failure of the equipment occurred due to lack of identification and subsequent rectification of deteriorating asset conditions. As components in a process plant operate, degradation is obvious and continues until finally resulting in a complete breakdown. Failure to detect the asset conditions that indicate a high likelihood for loss of containment can also result in disaster. With these regards, it is inferable that every incident starts in conjunction with the faulty assets operating in a process facility. The assets either lack adequate maintenance or improper operation originates the failure scenarios. The inadequate attention to the assets' health resulted in the following cases:
Case one:On April 8, 2004, at the Giant Industries' Ciniza oil refinery, Jamestown, Mexico, United States, mechanics were mistaken regarding the position of the valve wrench indicator while reinstalling a pump after repair. The consequence was a sudden release of flammable liquid. Subsequently after about 30 to 45 seconds of the initial release of hazardous Alkylate, fire and the first of the several explosions occurred. The incident injured six employees and caused the evacuation of non-essential employees.
Refinery equipment and support stnlctures were damaged and the production was not
resumed until the end of 2004. The review of the repair work prior to this incident revealed a history of repeated pump failures and showed the Giant's approach of following break-down maintenance instead of identifying the root causes of rrequent failure. At the same time, the LOTO procedure and the valve position indicator were also neglected during maintenance, which resulted in this undesired event (CSB, 2005).
Case two: On the morning of November 19, 1984, a pipe used to transport light hydrocarbons from a refinery to a storage terminal in Mexico City, Mcxico, ruptured and an accident occurred. Corrosion had gradually weakened a certain portion of the pipeline.
The light hydrocarbons quickly found an ignition source, triggering a series of fires and explosions, resulting in approximately 500 fatalities and destroying the LPG terminal.
The gradual degradation of the pipeline, which was either undetected or unaddressed, resulted in the failure of the equipment. This accident represents the largest series of major BLEVEs, and high fatality occurred because the housing was too close to the plant area. In this case also, due to the lack of attention towards maintaining asset integrity through routine inspection, and subsequent protection measure resulted in a catastrophic incident (Mannan, 2005).
Case three:On May 10,2008, the largest LPG producer of Indonesia, Balongan LPG Plant, had a major accident and had to shut down the plant for eighteen days for repair purposes. The accident occurred because of critical failure in a fluid catalytic cracking unit, which is a high pressure system. Consequences of this failure were significant, as the plant supplies around 30% of LPG to national market. The company was in the excess of twelve million US dollar production loss. The Health and Safety Executive, UK,
indicated the inability to predict or inability to anticipate in-service damage as one of the dominant root causes of failure in pressure systems. Thus, this un-anticipated in service damage of LPG process plant critical equipment can be considered as a failure of the asset integrity (Clough, 2009).
Case four: On March 23, 2005, a tire and explosion occurred at BP's Texas City Refinery in Texas City, Texas, killing 15 workers and injuring more than 180 personnel.
The Chemical Safety Board (CSB) identified several aspects of importance in this event related to poor asset integrity. The incident occurred in an isomerisation unit in start-up mode, with a number of important level instruments defective and some operations' experience gaps. This eventually led to overfilling of a distillation column, and liquid overflowed into a relief system that was directed to an atmospheric vent in the unit. The vent system also filledwithliquid,andeventuallygasolineoverflowedfromtheventinto the atmosphere of the process area. The failure to take effective emergency action resulted in a loss of containment incident. The vented gasoline certainly found an ignition source and a vapor cloud explosion occurred. The investigation reports of both Mogford (2005) and Baker (2007) panel pointed out asset integrity related several other underlying issues as the cause of the accident.
Ca.l'ejive:Beyond the process industries, on May 25, 1979, a DC-IO crashed on takeofTat the Chicago's O'Hare Airport when the pylon holding the left engine to the wing failed.
The resulting crash killed 273 people including 2 on the ground. The damage to the pylon was a result of incorrect maintenance procedures during the replacement of some internal bearings eight weeks before the crash. Ignoring the standard procedure of removing the
engine prior to the removal of the engine pylon, both the engine and pylon were removed atonetimeanda forklift was used to hold it in place. A failure of the forklift's hydraulic system left the engine unsupported and damaged the pylon. The damage went unnoticed for several flights, getting worse with each flight. Finally, during the incident, the pylon failed and tore the left engine away fi'om the wing (CCPS-RPPS, 2007).
These are only a few brief examples of occurred incidents in different areas directly related to lack or failure of asset integrity issues/concerns. Beyond these, there arc also several accident scenarios in the hydrocarbon industries that arc listed in "The 100 Largest Losses 1972 - 2009" and compiled by Clough (2009). The undesired incidents in the process facility arc some portentous signs for ncar future serious mishaps. The investigation of accidents/incidents in process industries revealed that in most oflhe cases the root causes of the incident were related to the negligence of asset integrity assurance or poor asset integrity systems. These incidents are occurring routinely one after another, and the desired integrity of asset has yet to be achieved. Much more attention is required to maintain the integrity level of the process plant.
Besides the accident scenario, the annual unwanted downtime in North American industry causes production loss of more or less 5% of total production, which is equivalent to staggering US$ 20+ billion annually. The numbers show the impact of downtime on overall performance and become a threatening issue for the survival of an industry. Much of this can be attributed to the failure of the industry to maintain the integrity of the assets or lack of recognition for necessary asset integrity. On the other hand, poor pertonnanee
of asset integrity runs counter to the basic objective of industry being able to operate reliably while avoiding unwanted scenarios.
Over the last several decades, substantial improvements in the industry have been observed in the area of lost time injury frequency (L1TF) and total recordable incident rates (TRIR), as shown in the Figure I-I (OGP, 2010). But, satisfaction with good occupational health and safety performance docs not ensure the occurrences of serioLIS mishaps in the future. The recent undesired incidents in the oil and gas sectors arc some portentous signs for ncar n.lture severe accidents. The anatomy of Texas City (2005) incident reveals that overlooking to asset health condition, i.e. the lack of mechanical, operational, and personnel integrity, were primarily responsible for the occurrence of the incident. This enforces the requirement of asset integrity, which had been neglected over the years. Again, requirement of asset integrity, in a process facility bccome more dominant with the increasing life of assets. With the ageing condition of equipment in a proccss plant, degradation progresses at a faster rate than expected. This causcs frequent failurescenarios,andplantdowntimealsoincreasesasaconsequenceofotherincidents.
Engineering structures, equipment, safety systems and components playa vital role in the process industry in fulfilling business requirements. Any threat to these components will also threaten the performance of overall assct integrity. At the same time, most of the process industry deals with hazardous materials, and loss of containment of these could be catastrophic. So, the requirement of asset integrity is two-fold: one is for keeping the equipment in operating condition and another is keeping the hazardous material inside the containment. This could be achieved by ensuring asset integrity. Engineering integrity is
an integrated system in which every component anccts other component in overall system. So process industries should be aware that failure to maintain the integrity of any asset could have potential effects on humans, environment, and even on the financial aspect of the industry.
Totolrecordobleiniuryrote-compony&controclors lo.sttimeinju(yfrequency-c.ompony&cOl\lroctors p~r'f1"f1on hOUfJ lVorked pe' 'Tlil/!on hours woro:ed
o2000 20012002 2003 2004 2005 2006 2007 2008 2009 02000 ](J()l 2002 2003 200-:~OO52C06 2007 2008 2009
Figure I-I: (a) Total recordable incident rate (b) Lost time injury frequency (OGP, 2010) Asset integrity refers to the strategies and activities intended for maintaining plant assets or equipment to ensure that they remain available, safe, and reliable in order to operate continuously. It includes characteristics such as design, operations, maintenance, and inspeetionpropertytomaximizereturIl from operating assets. The importance ofertcctive asset integrity increases as the industry assets continue to age. This issue has been realized by the Offshore Division, HSE (2007), for the offshore installations of UK continental shelf. Realizing the requirement of improving the integrity of installations to overcome the risk of major accidents, they have initiated the KP3-Asset integrity
program.Inaddition, when there is effeetive asset integrity, industries will have safer process plants with less accident, fewer leaks, and less damage to thc environment.
Eventually, this will cnhance the reputation of the organization. By implementing an effective asset integrity maintaining stratcgy, industries will significantly reduce serious damage to human livcs and to the environment. As well, industries will even have improved business performance.
To assure integrity of assets, most of the organizations involved consultancy companies and addressed some particular safety critical components to determine the overall plant characteristics. Sometimes organizations carried out only the inspection and assess the overall asset performance, ignoring the maintenance activity, human factor, and organizational issucs. On the othcr hand, to determine the pcrformancc of the plant organizations usually rely on the occupational health and safety performance. So, eventually these attempts turned out to be inadequate for maintaining and monitoring asset integrity and required a comprehcnsive approach. For ensuring asset integrity, a holistic approach is to be developed and followed that will consider cvery aspect of asset related issues. All threatening aspects of asset integrity should be neutralized proactively for the target of an incident frec facility.
Monitoring the performance of asset integrity is one of the most important and challenging issues in the assct integrity management program. Integrity monitoring should be fact-based, rather than opinion based, and may includc the following strategies pointedoutbyOGP(2008):
Key performance indicators (KPI), or simply performance indicators
ii. Barrier performance standard verification iii. Audit findings
iv. Incident and accident investigations
Benchmarking and lessons learned from external events
In this research, for the purpose of monitoring, reviewing and evaluating, the asset integrity indicator system is adopted. Catastrophic or major incidents due to loss of assel integrity in process plant arc relatively rare but not completely avoidable. That is why it is important to monitor asset performance and record even minor incidents which will eventually ensure the integrity in process facilities. But, for a process facility the indicators for monitoring asset integrity could be of quite large numbers as to cover every aspect. Furthermore, the information through the indicators could also be in varying characteristics and importance levels. As a result, it would not be possible, neither practical, to use all these parameters as indicators for asset performance monitoring or assessing. Therefore, in this research asset integrity indicator system will adopt a risk- based approach. The selection ofOoor level indicators will be based on the characteristics of risk associated with the events related to assets. The risk-based indicator system can simplify the complex array of information related to asset integrity. The consideration of risk characteristics will also allow the appropriate quantification of the indicators outcomes and numerical figures can be obtained for further aggregation.
1.2 Asset
An asset in respect to process industry is considered as an engineered piece ofequipment that is essential for the overall function of a process industry and critical to every
industry's performance. According to Sutton (2010), assets arc all equipment, piping, instrumentation, electrical systems, and other physical items in a process unit. In a single word an asset is a physical facility that is required for process opcration and has distinct value to the organization. BSI PAS 55-1 (2008) defined asset as "Plant, machincry, propcrty, buildings, vehicles andother itemsthathave a distinctvalue tothc organization". So, for the process industries, which run three hundred sixty five days a year, seven days a week, and twenty tour hours a day, the need to upkeep the asscts condition is of prime importance. Thus, management of assets' is the highcst priority for the performance and growth of the industry. A physical asset can be considered as a critical factor in achieving business goals. To maintain the comprchensiveness of the asset integrity approach, this thesis will consider the aspect of tangible as wcll as intangible assets.
1.3 Asset Integrity
CCPS-RBPS (2007) express that the primary objective of the asset integrity clement is to help ensure reliable performance of equipment designed to contain, prevent, or mitigate the consequences of a release of hazardous materials or energy. Searching through the literature and different regulatory organizations' guidelines resulted in identifying five major types of asset integrity, defined as follows:
HSE (2007) defined"asset integrity as the ability of an asset to perform its required function eftcctivelyand efficiently whilst protecting health, safety and the environment."
On the other hand CCPS- RBPS (2007) also definedasset integrity in the same manner:
"The asset integrity clement is the systematic implementation of activities, such as
inspcctionsandtcstsncccssarytocnsurcthatimportantequipmcntwillbcsuitablc for its intcndcd application throughout its lifc."
Again, OGP (2008) dcscribcd that "asset integrity is rclated to thc prcvcntion of major incidents.Itis an outeomc of good dcsign, construction and opcrating praetiecs.Itis achicvcd whcn facilitics are structurally and mcehanically sound and pcrfonn the process and produce thc products for which thcy were dcsigncd."
Thc CCPS (2010) guidclinc for proecss safcty mctrics dcfincd asset integrity as "work aetivitics that hclp ensurc that cquipmcnt is propcrly designcd is installcd in aecordancc with specifications, and rcmains fit for purposc ovcr its lifccyclc."
Finally, Piric (2007) of DNV dcfincd asset integrity as a "continuous proccss of knowlcdgc and cxpcricncc applicd throughout thc lifccyclc to managc thc risk of failurcs and cvcnts in dcsign, construction, and during operation of facilitics to cnsurc optimal production without compromising safcty, hcalth and cnvironmcntal requiremcnts."
From thc abovc dcfinitions, it can bc summarizcd that an asset in a proccssing facility achicvcs intcgrity when it opcratcs as dcsigncd, which mcans it is bcing opcratcd salcly following standard proccdurc with compctent pcrsonncl and complying with all ncccssary maintcnancc, inspcctions and tcsts; to be ablc to opcratc for its dcsigncd lifc mcans replaccmcnts, rcnovation, up-gradation, and rcpairs i.c. maintcnance, must be donc in a timcly, planncd manncr, conforming dcsign eodcs and cnginccring standards. For all asscts' associatcd risks to remain as low as reasonably practicablc, mcans all safeguarding and cmcrgcncy systcms associatcd with thcassct must bc in excellcnt shapc and able to handlc any risk cscalation situation or subscqucnt damage trom incidcnts.
10
This charactcrization of asset intcgrity will ultimatcly assist in dctcrmining thc cxtcnt of assct performance mcasurcmcnt through risk bascd indicators.
1.4 Factors Affecting the Integrity ofAsset
Tcchnical issucs havc thc grcatcst impact on thc intcgrity ofasscts. Othcr than tcchnical issucs, which arc highlightcd hcrc as a mcchanical intcgrity, opcrational, organizational, and pcrsonnel rclatcd issucs also havc substantial impact on thc assct intcgrity conccrn.
The following arc thc major mcchanical issues that havc thc utmost impact on thc intcgrityofassct opcrating in a proccss facility:
Extcrnal and intcrnal corrosion and crosion ofsystcms, structurcs and componcnts which is also rcsponsiblc for rcduction of componcnts' uscfullifc.
Fatiguc condition ofwcldcdjoints in systcms, structurcs and componcnts.
iii. Corrosion undcr insulation is a threatcning issuc that causcs juvcnilet~lilurcof componcnts.
iv. Inappropriate spccification, application, use, and maintcnancc of insulation and coating matcrials, as wcll as cathodic protection, contributc to corrosion.
Vibration levcl, ovcrprcssurc, over tcmpcraturc, ovcrloading situation bcyond design limit, and instrumcntation that monitor critical opcrational paramctcrs.
vi. Backlog of maintcnance resultingtromcxccssivc dcfcrrals, lack of tcchnical rcsourccs to conduct themaintcnancc, maintcnancc staffing,and lack of prioritizing tcchniquc for detcrmining safcty and cnvironmcntally critical cquipmcnt.
II
vii. Equipment design and selection, personnel competencies, and inspection strategy and maintenance planning and schedules.
viii. Ageing of operating assets. As ageing facilities approach their designed life, management also reduces maintcnanee costs as production levels decline, which in turn contributes to an increased risk of major accidents.
Sclcction of spare parts and consumables formaintenanceandopcration.
Meteorological phcnomcnon can also affect thcavailabilityofassets.
Besides thcsc, the issues related to operational and personnel activities that have most impactontheperformanceofassctintegrityare:
i. Incomprehensivencss of operating instructions and often continued operation beyond the safe design operating limits.
Management of change issues arc not executed following guidelines and not communicated properly.
iii. Immature safety culture and lack of management commitmcnt and support lor ensuring safety performance.
Poor integration between maintenance and operations' management systems.
Risk management strategy and lack of root cause analysis to determine the issues that led to an incident or failure.
Human factors including deliberate damage and competency of plant personnel.
vii. Poor communication system.
viii. Lack ofadequate technical and interpersonal trainings.
12
1.5 Why Need Performance Measurement?
The famous industrial revolutionary Peter Drucker said, "It is not possible to manage what you cannot control and you cannot control what you cannot measure." Rouhiainen (1990) also realizes the importance of measurement irrespective of objectives and suggests that "Measuremcnt is an absolute prerequisite for control, whethcr this is the control of production quality, accidents, or any other component of an industrial system."
Again Amaratunga et el. (2002) defined mcasurcmcnt as something that providcs "thc basis for an organization to assess how well it is progressing towards its predetermined objectives, helps to identify areas of strengths and weaknesscs, and dccides on fulure initiatives, with the goals of improving organizational performance." Performance measures refers to an indicator scheme used by management to measure, report. and improve performance and arc classed as either a key result indicator, a performance indicator, or a key performance indicator (Parmenter, 2007). The measurement of performance is important because of the following issues:
Identification of the current performance gap with the desired performance.
For managing strategies, executing initiatives, and evaluating performance.
iii. Indication ofprogress towards closing the gap between desired and outcome.
iv. For effective and efficient control of the equipment reliability for its purpose.
To ensure current performance is broadly communicated and thoroughly understood by different levels of management.
vi. Performance improvement through the involvement of multi level management.
13
vii. To enable a proactive management environment along with reactive management.
Fortunately, catastrophic accidents occur on a relatively infrequent basis. However, when they do occur, they usually involve a lot of investigation and root cause analysis activity.
The investigation rcvcals a number of for the accident secnario, which helps other stakeholders to learn from those situations. Other than wait for an accident to occur and investigate to identify the causes of failure, the assets' real time performance should be monitored. Otherwise, every process facility will be in need a of very robust and unnecessary accident protection system. The performance monitoring should be based on the risk based indicator system. Risk is inherent in all aspects of the asset maintenance and operation. Hence, the control of risk is central to asset integrity. This risk based monitoring of asset performance will cover both the active and reactive monitoring aspect. The rule thumb from Pareto charts states that 20% of equipment represents 80%
of the risk, so the idea is to focus on that 20% of equipment (API, 2000). To ensure the asset integrity, this 20% of equipment should be given more importance and can be categorized as most critical or highly risk significant components. Thus, risk-based performance monitoring will allow problems to be identified and corrective action to be taken before a serious incident occurs.
1.6 Performance Measure
Performancc measures constitute the 'Check clement' of the famous Deming's Plan-Do- Check-Act (PDCA) cycle. The Check clement involves monitoring different activities and strategies, as well as determining thepcrformance gap between current and expected 14
outcomes. Over the years, different types of performance measuring tools have been used depending on the area of measurement and the objectives ofmcasurcmcnl. Thcrc are four major types of performance measures for an organization, including: input to the system measure, process activity measure, and output or outcome measures. Among these.
OECD (2008) guidelines deal with thc activity and outeome measures using indicators, for performance measurement. Parmenter (1997) mentioncd that thcre are three types of pcrformancc measures, and thcseare: key result indicators, pc rformance indicators, and key performancc indicators. KRls rcveals how you have done in a perspective, Pis direct towards what to do, and the KPls indicatc what to do to increase performance dramatically.Parmenterusesanonionanalogytodescribetherelationshipofthesethree measures, as shown in Figure 1-2. The outside skin describes the overall condition of the onion, the amount of sun, water, and nutrients it has received, as well as how it has been handled from harvest to supemlarket shelf. However, as layers are peeled off the onion, more information is found. The layers represent the various performance indicators, and the core represents the key performance. KPls represent a set of measures focusing on the aspects of organizational performance that are the most critical for the current and future success of the organization.
15
Figurcl-2: Thrcc typcs ofpcrformancc mcasurc (D. Parmcntcr, 1997) Ovcrall cquipmcnt cffcctivcncss (OEE)
Thc OEE conccpt is usually utilizcd to mcasure thc efTcetivcncss of a manufacturing process, but it can also bc utilized in non-manufacturing operations. akajima (1988) dcfincd OEE as "a bottom-up hicrarehy approach whcrc an intcgratcd workforcc strivcs to achicvc ovcrall cquipmcnt cffcctivcncss by climinating thc six big losscs." Godfrcy (2002) cxplores thc bcnefit of using OEE to inform decision making throughout thc litceyelc of an assct along with thc power of OEE mcasurc to improvc thc opcrational pcrformance." Thc overall pcrtormance for a singlc componcnt or for an cntirc filcility can bc measurcd dcpcnding on thc cumulativc impact of the thrcc OEE factors. OEE is a mcasurc of tota) cquipmcnt pcrformancc i.e. thc dcgrcc to which thc assct is doing what it is supposed to do basc on OEE dimcnsions: actual availability, pcrformance cfficicncy, and quality of product or output. Thus, OEE is considcrcd a kcy factorinmcasuring both productivity and cffcctivcncss, and thc hierarchy ofmctrics focuscs on how cffectively a
16
manufacturing opcration is utilizcd. OEE mcasurcmcnt is also commonly uscd as a KPI in conjunction with lean manufacturing cfforts to providc an indicatorofsucccss (Stamatis, 2010). Yct it is not a statistically valid tool and not intcndcd for usc as a corporatc or plant lcvcl mcasurc. OEE pcrformancc pcrccntagc assumcs that all cquipmcnt-rclatcd losscs arc cqually important andis a rough cstimation of sclectcd cquipmcnt cffcctivcncss.
1.7 AssetIntegrity Indicator
Thc tcrm 'indicator' traccs back(0thc Latin vcrb 'indicarc', mcaning to disclosc or point out, to announcc or makc publicly known, or to cstimatc or put a pricc on (Hammond, 2005). In accordancc with thc dcfinition of Building Tcrms fi'om Standards Australia (BTSA) (SAA HB50, 1994), asset integrity indicator can bc dcfincd as "a qualitativc or quantitative mcasurcofthcqualityofthcassct's pcrformancc, cfficicncy, productivity of an activity which cnablcs a comparison to bc madc for managcmcnt proccss of pcrformancc against a standard targct." Again, in glossary ofkcy tcrms in cvaluation and rcsults-bascdmanagcmcnt, OECD (2010) dcfincd an indicator solcly as a "quantitativc or qualitativc factor or variablc that providcs a simplc and rcliablc means to mcasurc achicvcmcnt, to rcflcct thc changcs connected to an intcrvcntion, or to hclp asscss thc pcrformancc ofa dcvclopmcnt actor." This spccific mcaning is uscd to clarify conccpts and diminish tcrminological confusion (OECDIDAC, 2010). EEA (2005) also dcfincd an indicator as a quantitativc mcasurc that can bc uscd "to illustratc and communicatc complex phcnomcna simply, including trcnds and progrcss ovcr time." HSE, KP3 rcport
17
(2007) stated that indicators measure performance and provide feedback on what is happening so that the user can shape the appropriate actions to respond to changing circumstances. Indicators have a variety of options in measurement, and an organization has to choose activities that arc related to their goal. Therefore the identified indicators should be established and formulated to fi.dfill the overall goal. According to British Standard (2005), indicators allow an organization to performLhe following activities:
i. Measure Lhestatus Evaluate the performance iiI. Compare performance
Identify strengths and weaknesses Set objectives
vi. Plan strategies and actions
vii. Share the results in order to inform and motivate people and viii. Control progress and changes over time.
1.7.1 Purpose of asset integrity indicators
Indicators became essential, as well as effective, tools for tracking asset integrity performance in process industries. Indicators that correspond to asset integrity have several advantages that encourage their usc for asset performance measurement. Target oriented appropriate indicators also act as a source of asset management information.
CCPS-RBPS (2007) guidelines also enforce the requirement ofmetrics that could be used Lo monitor asset integrity. The following benefits of indicators arc a few rcasons they should be used for performance measurement:
18
Playa crucial role in making asset management information system operational.
Reduce time for locating the fault in assets and locating the latent weakness of operatmgassets.
iii. Identify the early signals of deteriorating asset performance that could underpin the asset integrity.
iv. Provide warning of approaching troublc bcforca scrious incident occurs.
Allow an ease in investigation and root-cause analysis through warning system so thatcorreetive action can be taken betoreany unwanted incident occurs.
vi. Provide guideline to management for rational decision making in maintenance prioritizationandtoachievetop-Ievelpolicymakers'atlention.
vii. Compare and tune of target performance with theaclual performa nce.
viii. Identify strong and weak areas of performance and knowledge transfer Ii·om similar strong and weak areas.
ix. Act as powerful motivational tools that provide an ease in decision making.
Asset integrity indicators should provide the earliest possible warning of declining performance that could be increasing the operational risk.Itis theretoreessential to use a broad set of indicators to cover process plant general performance in the region of maintenance, operation, and manpower related activities.
1.7.2 Types of Asset Integrity Indicators
Surprisingly, over the years, proeess industries around the world were sat is tied with measurement provided through lagging indicators. More specifically, they werc rclying on occupational safcty indicators, such as thc fatal accidcnt rate (FAR), lost time injury 19
fi'cqucncy (LTIF), and total rccordable incident ratc (TRJR). Managcmcnts arc imposing morc cmphasis only on improving thc forcsaid lagging catcgory in ordcr to provc thcir systcm are opcrating cfficicntly, whilc neglccting the plant's physical assct condition.
Mogford's (2005) invcstigation rcport on Tcxas City cxplosion pointcd out that thc sitc has numcrous mcasurcs for tracking opcrational, commercial, cnvironmcntal, and safety pcrformancc. But, thcsc indicators, mostly of lagging typc, not prioritizcd and did not clcarly focus on thc Icading indicators as wcll. Mogford concludcd that "by definition, catastrophic and major proccss incidcnts arc rarc cvcnts, and performancc mcasurcs nccd to bc preferably focuscd on Icading indicators, or at Icast lagging indicators ofrclcvant, morc '"j'cqucnt smallcr incidcnts." Thc samc issuc was also idcntilicd in thc Noradic Nuclcar Safcty Rcscarch projcct rcport (Laakso et aI., 1994), which statcd lhat throughout thc opcration of nuclcar powcr plant only a fcw major safcty significant dircct cvcnts can occur. So, with thc Iimitcd quantity ofdircct evcnt information, managcmcnts havc littlc todctcrmine futurecsscntials. Thiscnforcesthcrcquircmentofcasilymcasurablcindircct plant pcrformancc paramctcrs that will also provide an advanccd warning of dccaying pcrformancc. In this way, thc dircct impact can also be avoidcd as wcll. Bakcr (2007) statcd thc importancc of use of lagging and leading indicators as rcactivc and activc monitoring of performance, rcspcctivcly,where "reactive monitoring allows an organization to idcntify and corrcct dcficiencies in response to spccific incidcnts or trcnds and active monitoring cvaluatcs thc prcscnt state of a facility through the routinc and systcmatic inspection and tcstingofwork systems, premiscs, plant, and cquipmenl."
20
In the updated guideline, HSE, UK (2006) introduced the concept of 'dual assurance' with both leading and lagging indicator utilization for ultimate risk control. Ifboth types of indicator sets arc employed in a structured and systematic way then it will ensure the cflcctiveness of critical risk control system. BSI PAS 55 (2008) described the requirement of proactive, reactive, leading, lagging, quantitative and qualitative measure for physical asset. IAEA (2000) also pointed out that monitoring pcrformance with combination of Icading and lagging indicator sets provides the bcst performance measurcment system. Considering all of the above issucs, asset integrity indicators arc also categorized into the following two major groups:
Leading or proactive indicator Lagging or reactive indicator
The combined application of indicators for monitoring asset performance will provide a comprehensive view of asset condition. Based on the performance of leading indicators, the outcome can be predicted and, with the lagging indicators result areas for improvement in the leading inputs can be determined. In the long run, lagging performance will be improved on the basis of good performances of leading indicators.
1.7.3 Characteristics of Asset Integrity Indicators
The selection of effective indicators can be done after a complete and thoughtful revision and collaboration of key processes, equipment, organization culture, and activities involved in process facilities that possess grcater risks. The successfulness of assct integrity indicator system depends on the proper selection of indicators and their precise uses. If the indicators arc not selected correctly and used improperly then this could be 21
misleading rather than assisting in performance measurement. The selection of indicators should be based on certain characteristic that will assist in identifYing the proper indicators. To determine the appropriateness of potential indicators, Mc eeney (2005) provides a detailed set of criteria known as the 'SMART' test. The acronym stands for the live characteristics: specific, measurable, attainable, realistic, and timely. Identified indicators have to comply with these five characteristics for maintaining quality and elTeetiveness in performance measurement. IAEA (2000) also identified an ideal set of characteristics for selecting operational safety indicators for maintaining the quality of indicator information. Indicator characteristics varied with their context of application. In case of asset integrity where too Illany issues arc involved, effective characteristic selection is a major concern. To identifY generic sets of indicators for monitoring asset integrity, suggested characteristics of IAEA (2000) and DOE (2002) criteria were analyzed thoroughly to figure out required potential characteristics of asset integrity indicators. Analyzing these potential traits, the following characteristics arc preferred lor selecting asset integrity indicators that will go with the risk based concept too:
Relevance and direct relationship with the assessment category.
Unambiguollsand lInderstandableat each level.
iii. Reliable, Illeaningful, and easily integrated to asset related activities.
iv. Capable of expressing in quantitative terms and able to provide information timely.
Capable of representing the risk significant issues involved in the operation.
22
The selection of indicators for assessing the asset integrity is a vital issue that determines the etTectiveness of the risk based asset integrity indicator strategy.
1.8 Objectives of Research
Maintaining equipment fitness for purpose and ensuring safety systcms functionality when necessary is of paramount importance to process industries (CCPS, 2007).Ina typical day, maintenance will spend 40% of its time investigating the root causes of a problem (Gonzalez, 2005). This significant amount of time can be reduced by introducing indicators for monitoring assets continuously. Considering the above issue along with the strategy and purposes of ensuring asset integrity, the objectives of this research work arc
Develop a generic hierarchical fTamework to relate the top level strategy of ensuring asset integrity with the events occurring on the site floor.
Identify comprehensive sets of risk based leading and lagging indicators in the mechanical, operational, and personnel areas of asset integrity following developed hierarchical framcworkandusing the standard guideline.
iiI. Develop sets of questionnaires: one for standardization of the hierarchical fTamework indicators weight and another for collection of basic level risk information.
iv. Develop an aggregation technique to provide the same basis for both types of indicators' risk estimation and to determine the top level risk index.
23
Validate developed model by collecting, aggregating, and integrating information
anddetermine leading and laggingrisk indexto monitor asset integrity
performance.
1.9 Novelty of this Research
The major focus of this research is the development ofa risk-based indicators system to
assess the asset integrity in process industries. This research is unique in asset integrity
area since it utilizes the risk definition by selecting indicators, collecting risk information, and aggregating risk levels based on the highest associated risk of the indicator. This
approach considers both the Icadingand lagging aspects of indicators that arc quantifiable in terms of risk and can be easily mapped with the standard risk index scale to determine
the asset's condition. This study also proposed a comprehensive set of multi-level
indicators that are easy to establish in particular process facility. Depending on the availability of current features and future requirements, the indicators can also be
excluded or included, respectively in the identified indicator sets. This developed approach is a comprehensive, systematic, and integrated risk based asset integrity
indicator system where the physical asset integrity in the section of mechanical and operational activity can be built on the personnel integrity of every employee.
1.10 ThesisOutline
The thesis is comprised of six chapters. Each chapter of the thesis illustrates the distinct
aspect of asset integrity indicators to achieve stated objectives.
24
Chapter I addresses the background, elucidating the asset, asset integrity, and asset integrity indicator concepts and, objectives.Italso discusses novelty of the proposed research.
Chapter 2 presents a brief review of performance measurement frameworks. Also, it presents a review of guidelines and literature pertaining to the current research work, followed by limitations of these approaches. This chapter also includes goals and described the scope of current research work.
Chapter 3 provides a detailed description of the risk based asset integrity indicator methodology. It includes the dclimitation and development strategy of multilcvel hierarchical indicator framework approach for asset integrity. Furthermorc, in accordance with the framework structure, indicators at each level along with leading and lagging indicators at specific level were identified. It also discusses the indicator data aggregation, as well as data collection policy and standardization of multilevel indicator weights.
Chapter 4 presents a benchmark study that determines the feasibility and applicability of developed indicator systems in different process plants.Italso describes the means for questionnaire dcvelopment for data collection followed by data analysis, evaluation, and result discussion.
Chapter 5 represcnts an additional work that is very much rclated with the asset integrity assurance issue. Itformulates arisk based spareparts inventory management methodology that will fulfill the spare parts requirement during maintenance.
Chapter 6 concludes the research work by summarizing the potentiality of the approach, followed by overall discussion and recommendations on future research scope in this area.
25
